Data-driven Approaches for Formal Synthesis of Dynamical Systems

My research lies at the intersection of control theory, machine learning and formal methods. This paper presents part of the work developed so far within the scope of my PhD and suggests possible future research directions. Towards trustworthy computing, my research has focused on simplifying the designing pipeline of safe and reliable AI systems. I have worked on data-driven controller synthesis, i.e., the automated generation of control systems from a given high-level specification with theoretical guarantees of correctness. In this way, I analyze the satisfaction of properties in both episodic and continual settings. Moreover, in my research I provide correctness for satisfying specifications using different approaches including abstraction-based techniques, game-theoretic techniques, and model-free reinforcement learning.